[00:00:02] ok [00:00:04] heh [00:00:07] 50 then :) [00:00:10] Retries aren't expensive, just don't want them going forever when they won't succeed [00:00:19] And taking up an execution thread ;-) [00:00:36] *slaves* should retry forever [00:00:39] Since we want them consistent [00:00:46] But github? meh [00:01:29] no_justification https://gerrit.wikimedia.org/r/#/c/408468/ [00:02:11] no_justification some how when i type dzahn in the reviewer dialog, his name dosen't come up [00:02:33] but if i enter his full email and click add, it works (no autocomplete but add's him as a reviewer though) [00:02:45] Weird [00:03:05] i wonder if this is the account index playing up [00:03:29] we load accounts from the db when logging in, but i think other things use the index. [00:06:03] thcipriani hi, it seems there has been a test running for 7hours on https://integration.wikimedia.org/zuul/ [00:06:16] (reported earlier by tgr). [00:06:32] paladox: :( OK, will look after swat (if I can SWAT with it in this state) [00:06:39] thanks :) [00:06:51] You can still swat i think :) [00:14:32] no_justification: paladox: thcipriani: the stuck TemplateStyles patch is the int/str zuul thing [00:14:39] ah [00:14:52] I guess i will set it as unbreak again [00:15:22] 10Continuous-Integration-Infrastructure, 10Zuul: Exception while launching job: TypeError: 'int' object has no attribute '__getitem__' - https://phabricator.wikimedia.org/T186381#3947956 (10Paladox) p:05High>03Unbreak! Broke again [00:14:32] no_justification: paladox: thcipriani: the stuck Temp... [00:15:34] 10Continuous-Integration-Infrastructure, 10Zuul: Exception while launching job: TypeError: 'int' object has no attribute '__getitem__' - https://phabricator.wikimedia.org/T186381#3947958 (10Legoktm) This is currently affecting 393285,5 / TemplateStyles. Luckily it's in the test queue and not blocking all merge... [00:16:00] legoktm i think it only affect's patches that depend on other changes [00:16:09] i guess that's what they missed when fixing this [00:16:25] seems like it [00:16:47] legoktm upstream doin't support zuul 2 [00:17:00] so we will need to fix this at least our self's but push it upstream for v3 [00:17:06] Ugh :( [00:17:34] they even removed the 2.x branch making it hard to update to the last 2.x commit. [00:18:37] in getBasePath [00:18:38] self.number[-2:], self.number, self.patchset) [00:18:41] should self.number be put with [00:18:45] int(self.numer)? [00:18:54] oh str [00:18:59] no, str(self.number) [00:19:04] ah [00:19:22] what is self.number? [00:19:31] self is a change object? [00:19:32] thcipriani gerrit's change number [00:19:43] thcipriani: there's a traceback on the bug [00:22:20] 10Continuous-Integration-Infrastructure, 10Gerrit, 10Wikidata: Make DataTypes repository read-only - https://phabricator.wikimedia.org/T182566#3947966 (10demon) 05Open>03Resolved a:03demon Someone did this it seems :) [00:22:27] 10Continuous-Integration-Infrastructure, 10Zuul: Exception while launching job: TypeError: 'int' object has no attribute '__getitem__' - https://phabricator.wikimedia.org/T186381#3947969 (10Paladox) cc @hashar i think we need to cover self.number with str so it's str(self.number[-2:])? and str(self.number) [00:23:05] well looking at the __repr__ I guess the number from the task is 407165 which would explain the problem. [00:23:12] 10Gerrit, 10Patch-For-Review: Fix access rules for gerrit - mediawiki/extensions/LanguageTool - https://phabricator.wikimedia.org/T186410#3947972 (10demon) 05Open>03Resolved a:03demon [00:24:26] 10Gerrit, 10VPS-project-icinga2, 10User-Zppix: Labs/Icinga2 group ownership - https://phabricator.wikimedia.org/T182941#3947974 (10demon) 05Open>03Resolved a:03demon [00:25:22] thcipriani yeh, it's due to a api change in 2.14 [00:25:25] makes it an int [00:25:32] so we need to convert it to a string [00:26:28] makes sense. [00:27:46] 10Gerrit, 10Release-Engineering-Team (Someday), 10Cleanup, 10Wikidata, and 3 others: Mark extension-Wikidata & wikidata-build-resources on Gerrit as ARCHIVED - https://phabricator.wikimedia.org/T181838#3803897 (10demon) * Looks like CI is done above * Github repos deleted * Phab mirror marked inactive * Bo... [00:30:07] 10Gerrit, 10Repository-Admins: Can not change group membership in gerrit as a group member anymore - https://phabricator.wikimedia.org/T173337#3524539 (10demon) This isn't to do with single-user-groups plugin (that was what created pseudo-groups called user-FOO you could use in ACLs to assign a person. But tha... [00:31:22] 10Gerrit, 10Repository-Ownership-Requests: Review membership of "mediawiki" Gerrit group - https://phabricator.wikimedia.org/T168216#3947993 (10demon) 05Open>03declined We don't revoke access for inactivity and never have. As long as people leave on good terms, they're more than welcome back any time. [00:33:13] 10Gerrit, 10Social-Tools: Gerrit group creation request: Create group for Social-Tools - https://phabricator.wikimedia.org/T154078#3947998 (10demon) 05Open>03stalled [00:36:02] no_justification lol Qualcomm is still running gerrit 2.7 [00:36:28] 10Gerrit, 10Cleanup, 10GitHub-Mirrors, 10MediaWiki-extensions-General, 10Repository-Admins: Archive the ShoutBox extension - https://phabricator.wikimedia.org/T155365#3948002 (10demon) 05Open>03Resolved Done. [00:37:39] Hey, if it works :p [00:38:02] 2.7 and 2.8 were *good* releases. [00:38:03] Very stable [00:38:14] That's why we stayed on 2.8.x so long [00:38:24] heh [00:38:49] If it's internal, there's not a huge security risk either. [00:39:02] heh [01:47:37] PROBLEM - Free space - all mounts on deployment-fluorine02 is CRITICAL: CRITICAL: deployment-prep.deployment-fluorine02.diskspace._srv.byte_percentfree (<11.11%) [02:31:45] (03PS1) 10Legoktm: Generate coverage for TemplateStyles [integration/config] - 10https://gerrit.wikimedia.org/r/408480 [02:31:57] (03CR) 10Legoktm: [C: 032] Generate coverage for TemplateStyles [integration/config] - 10https://gerrit.wikimedia.org/r/408480 (owner: 10Legoktm) [02:33:22] (03Merged) 10jenkins-bot: Generate coverage for TemplateStyles [integration/config] - 10https://gerrit.wikimedia.org/r/408480 (owner: 10Legoktm) [02:33:59] !log deploying https://gerrit.wikimedia.org/r/408480 [02:34:05] Logged the message at https://wikitech.wikimedia.org/wiki/Release_Engineering/SAL [04:08:18] (03PS1) 10KartikMistry: Add apertium-rus-ukr and apertium-ukr debian jobs [integration/config] - 10https://gerrit.wikimedia.org/r/408483 [05:58:19] (03CR) 10Hashar: [C: 032] Add apertium-rus-ukr and apertium-ukr debian jobs [integration/config] - 10https://gerrit.wikimedia.org/r/408483 (owner: 10KartikMistry) [05:59:39] (03Merged) 10jenkins-bot: Add apertium-rus-ukr and apertium-ukr debian jobs [integration/config] - 10https://gerrit.wikimedia.org/r/408483 (owner: 10KartikMistry) [07:07:40] RECOVERY - Free space - all mounts on deployment-fluorine02 is OK: OK: All targets OK [07:29:11] 10Phabricator, 10Release-Engineering-Team (Kanban), 10Operations, 10Patch-For-Review, 10User-Elukey: Apache on phab1001 is gradually leaking worker processes which are stuck in "Gracefully finishing" state - https://phabricator.wikimedia.org/T182832#3948321 (10elukey) >>! In T182832#3947179, @Paladox wro... [08:25:55] 10Phabricator, 10Release-Engineering-Team (Kanban), 10Operations, 10Patch-For-Review, 10User-Elukey: Apache on phab1001 is gradually leaking worker processes which are stuck in "Gracefully finishing" state - https://phabricator.wikimedia.org/T182832#3948363 (10elukey) All apache httpd settings rolled bac... [08:33:57] 10Phabricator, 10Release-Engineering-Team (Kanban), 10Operations, 10Patch-For-Review, 10User-Elukey: Apache on phab1001 is gradually leaking worker processes which are stuck in "Gracefully finishing" state - https://phabricator.wikimedia.org/T182832#3948368 (10elukey) Next step in my opinion would be to... [08:34:52] about --^: let me know if you need anything else from ops, afaics the next step is to follow up with upstream but I might be missing something :) [09:40:16] hashar: can you look at lintian error at: https://gerrit.wikimedia.org/r/#/c/408264/ ? Thanks :) [09:40:36] kart_: sure [09:40:48] hashar: new errors seems recent.. [09:41:04] kart_: yeah I slightly changed the way lintian run on debian-glue jobs [09:41:23] previously we were always runniing the version from the host (Jessie) [09:41:30] now it uses the lintian version from the target distribution [09:41:48] aha. Right. so, jessie-wikimedia will fail. [09:41:50] E: apertium-ukr changes: bad-distribution-in-changes-file jessie-wikimedia [09:41:52] hmm [09:43:01] https://wikitech.wikimedia.org/wiki/APT_repository says we use jessie-wikimedia [09:43:38] I gotta find an old build and check the output [09:44:00] but yeah the issue is the .changes file has: [09:44:05] Distribution: jessie-wikimedia [09:44:06] ( https://integration.wikimedia.org/ci/job/debian-glue/1015/artifact/apertium-ukr_0.1.0%7Er82563-1+wmf1+0%7E20180206061821.1015+jessie+wikimedia%7E1.gbp18d3d0_amd64.changes/*view*/ ) [09:44:58] but puppet package_builder has ./modules/package_builder/files/lintian-wikimedia/main/data/changes-file/known-dists [09:45:15] which has our custom distributions in it [09:45:24] guess I should copy that file in the cow build environement [09:50:40] 10Continuous-Integration-Config, 10Packaging: debian-glue lintian fails with: changes: bad-distribution-in-changes-file jessie-wikimedia - https://phabricator.wikimedia.org/T186587#3948455 (10hashar) [09:51:52] (03CR) 10Hashar: [C: 032] "I have deployed apps-android-wikipedia-npm-node-6-docker yesterday and that fixed the issue." [integration/config] - 10https://gerrit.wikimedia.org/r/408376 (https://phabricator.wikimedia.org/T185931) (owner: 10Hashar) [09:52:36] 10Continuous-Integration-Infrastructure, 10Release-Engineering-Team (Kanban): npm-node-6-docker tests failing for Android project (test fallback to broken npm 1.4.21 instead of 3.8.3) - https://phabricator.wikimedia.org/T185931#3948467 (10hashar) [09:53:31] 10Continuous-Integration-Infrastructure, 10Release-Engineering-Team (Kanban): npm-node-6-docker tests failing for Android project (test fallback to broken npm 1.4.21 instead of 3.8.3) - https://phabricator.wikimedia.org/T185931#3928972 (10hashar) Seems it is working properly now. I am not sure what is mobile t... [09:57:18] kart_: filled and I am looking for a proper fix [10:05:45] 10Phabricator (Upstream), 10Developer-Relations, 10Legalpad, 10translatewiki.net, and 4 others: Use translatewiki.net to localize Phabricator - https://phabricator.wikimedia.org/T225#3948497 (10Petar.petkovic) [10:13:24] hmm [10:14:47] kart_: fixed. But really I have no idea how :] [10:15:01] Looking.. [10:21:29] PROBLEM - Mediawiki Error Rate on graphite-labs is CRITICAL: CRITICAL: 40.00% of data above the critical threshold [10.0] [10:26:32] RECOVERY - Mediawiki Error Rate on graphite-labs is OK: OK: Less than 1.00% above the threshold [1.0] [10:33:58] PROBLEM - Puppet errors on deployment-etcd-01 is CRITICAL: CRITICAL: 11.11% of data above the critical threshold [0.0] [10:42:09] kart_: I screwed up my change sorry [10:44:44] hashar: no worries. There is no hurry to deploy those packages :) [10:51:06] (03PS3) 10Hashar: debian-glue: run lintian as a hook [integration/config] - 10https://gerrit.wikimedia.org/r/408268 (https://phabricator.wikimedia.org/T186494) [10:51:06] (03PS1) 10Hashar: debian-glue: use lintian wikimedia profile [integration/config] - 10https://gerrit.wikimedia.org/r/408522 (https://phabricator.wikimedia.org/T186587) [10:53:22] * hashar watches https://integration.wikimedia.org/ci/job/debian-glue/1019/console [11:03:44] Cannot find profile wikimedia (in /usr/share/lintian/profiles) at /usr/bin/lintian line 979. [11:03:45] bah [11:04:29] 10Project-Admins, 10I18n: Add "Turkish-sites" tag - https://phabricator.wikimedia.org/T186403#3948584 (10Liuxinyu970226) >>! In T186403#3945746, @Aklapper wrote: > Has the/a Turkish community requested interested in having such a tag to organize their work, or how would it be used by who? > I'd like to unders... [11:11:53] (03PS2) 10Hashar: debian-glue: use lintian wikimedia profile [integration/config] - 10https://gerrit.wikimedia.org/r/408522 (https://phabricator.wikimedia.org/T186587) [11:13:12] (03CR) 10Alexandros Kosiaris: [C: 04-1] "The if clause looks wrong, the rest looks fine" (031 comment) [integration/config] - 10https://gerrit.wikimedia.org/r/408522 (https://phabricator.wikimedia.org/T186587) (owner: 10Hashar) [11:13:17] kart_: I think I got it working properly https://integration.wikimedia.org/ci/job/debian-glue/1023/console [11:13:30] the trick is to bindmount ema lintian profile in the build env using USER_BINDMOUNTS='/usr/share/lintian/profiles/wikimedia /usr/share/lintian/vendors/wikimedia' [11:13:39] which exposes those directories to the build env [11:13:47] then instruct lintian to use that "wikimedia" profile [11:14:03] there are still a few warnings left, but they are not important [11:16:26] (03CR) 10Hashar: debian-glue: use lintian wikimedia profile (031 comment) [integration/config] - 10https://gerrit.wikimedia.org/r/408522 (https://phabricator.wikimedia.org/T186587) (owner: 10Hashar) [11:19:48] (03PS3) 10Hashar: debian-glue: use lintian wikimedia profile [integration/config] - 10https://gerrit.wikimedia.org/r/408522 (https://phabricator.wikimedia.org/T186587) [11:22:43] (03CR) 10Alexandros Kosiaris: debian-glue: use lintian wikimedia profile (031 comment) [integration/config] - 10https://gerrit.wikimedia.org/r/408522 (https://phabricator.wikimedia.org/T186587) (owner: 10Hashar) [11:27:37] (03CR) 10Alexandros Kosiaris: [C: 04-1] debian-glue: use lintian wikimedia profile (031 comment) [integration/config] - 10https://gerrit.wikimedia.org/r/408522 (https://phabricator.wikimedia.org/T186587) (owner: 10Hashar) [11:34:33] 10Phabricator, 10Release-Engineering-Team (Kanban), 10Operations, 10Patch-For-Review, 10User-Elukey: Apache on phab1001 is gradually leaking worker processes which are stuck in "Gracefully finishing" state - https://phabricator.wikimedia.org/T182832#3948659 (10Paladox) @elukey do you mean phabricator ups... [11:34:36] 10Project-Admins, 10I18n: Add "Turkish-sites" tag - https://phabricator.wikimedia.org/T186403#3948660 (10Aklapper) @Liuxinyu970226: It does not matter who has Turkish knowledge or not in this case... It matters who will actually use the tag and how. I do not want categorization for the sake of categorization.... [11:36:17] 10Project-Admins: Add "Turkish-sites" tag - https://phabricator.wikimedia.org/T186403#3948661 (10Aklapper) Unrelated to #i18n hence removing [11:36:37] 10Phabricator, 10Release-Engineering-Team (Kanban), 10Operations, 10Patch-For-Review, 10User-Elukey: Apache on phab1001 is gradually leaking worker processes which are stuck in "Gracefully finishing" state - https://phabricator.wikimedia.org/T182832#3948663 (10elukey) >>! In T182832#3948659, @Paladox wro... [11:38:20] I don't like the new Gerrit UI. I prefer what we have now :) [11:48:45] 10Continuous-Integration-Infrastructure, 10Operations, 10Patch-For-Review: legoktm can't deploy docker images on contint1001 - https://phabricator.wikimedia.org/T186475#3948703 (10akosiaris) 05Open>03Resolved a:03akosiaris I am thinking this is resolved now. Don't forget to logout + login to make sure... [11:53:56] RECOVERY - Puppet errors on deployment-etcd-01 is OK: OK: Less than 1.00% above the threshold [0.0] [12:11:50] 10Project-Admins: Add "Turkish-sites" tag - https://phabricator.wikimedia.org/T186403#3948727 (10Amire80) Well, this is related to i18n in the sense that a lot of issues that are related to Turkish //language// are related to internationalization by their nature. Common problems for computing in Turkish is the c... [13:11:28] 10Phabricator: My username shows wrong - https://phabricator.wikimedia.org/T185998#3948802 (10Aklapper) a:03CodeCat [13:14:17] 10Project-Admins: Create a project for Tool-Hashtags - https://phabricator.wikimedia.org/T186103#3948804 (10Aklapper) 05Open>03Resolved a:03Aklapper I think we could still rename to `Hashtags Search` if wanted or if there is ever a collision? Project created: https://phabricator.wikimedia.org/project/view... [13:24:40] 10Phabricator (Upstream), 10Upstream: Update Phabricator OAuth scopes in documentation - https://phabricator.wikimedia.org/T185447#3948826 (10Aklapper) [13:37:38] hasharAway hi, im wondering if you could take a look at this UBN task https://phabricator.wikimedia.org/T186381 please [13:37:41] when your around [13:38:22] no_justification ohh https://gerrit.wikimedia.org/r/plugins/gitiles/ :) [14:10:21] 10Continuous-Integration-Infrastructure, 10Release-Engineering-Team (Kanban): npm-node-6-docker tests failing for Android project (test fallback to broken npm 1.4.21 instead of 3.8.3) - https://phabricator.wikimedia.org/T185931#3948879 (10Dbrant) 05Open>03Resolved For technical stuff like this, I can still... [14:20:59] 10Gerrit, 10Release-Engineering-Team: Groups in gerrit are really slow - https://phabricator.wikimedia.org/T186511#3948909 (10Paladox) 05Open>03declined [14:22:06] 10Gerrit, 10Patch-For-Review: Setup gitiles for use with gerrit as a repo viewer - https://phabricator.wikimedia.org/T184116#3948910 (10Paladox) gitiles is now installed and can be accessed at https://gerrit.wikimedia.org/r/plugins/gitiles/ [14:26:02] no_justification ohhh there's a browse link in the diff view that takes you to gitiles [14:26:02] https://gerrit.wikimedia.org/r/plugins/gitiles/operations/puppet/+/refs/changes/65/407465/7/modules/apt/files/apt-upgrade.py [14:27:58] though the url is broken in polygerrit [14:27:58] heh [14:28:22] * paladox looks for a fix [14:33:12] no_justification it's also so much easier to debug polygerrit, as it includes the template name like gr-diff-view [14:35:22] gr-patch-range-select [14:38:01] no_justification ah found the fix, someone fixed it in gitiles master branch [14:38:06] so im backporting it here https://gerrit-review.googlesource.com/#/c/plugins/gitiles/+/157731/ [14:40:47] 10Project-Admins: Add "Turkish-sites" tag - https://phabricator.wikimedia.org/T186403#3943170 (10Aklapper) 05Open>03Resolved a:03Aklapper Thanks for elaborating. Let's try. Requested public project #Turkish-Sites has been created: https://phabricator.wikimedia.org/project/view/3231/ Please encourage inte... [15:12:29] (03CR) 10Hashar: debian-glue: use lintian wikimedia profile (031 comment) [integration/config] - 10https://gerrit.wikimedia.org/r/408522 (https://phabricator.wikimedia.org/T186587) (owner: 10Hashar) [15:12:37] (03PS4) 10Hashar: debian-glue: use lintian wikimedia profile [integration/config] - 10https://gerrit.wikimedia.org/r/408522 (https://phabricator.wikimedia.org/T186587) [15:12:45] (03CR) 10Hashar: [C: 032] debian-glue: run lintian as a hook [integration/config] - 10https://gerrit.wikimedia.org/r/408268 (https://phabricator.wikimedia.org/T186494) (owner: 10Hashar) [15:12:53] (03CR) 10Hashar: [C: 032] debian-glue: use lintian wikimedia profile [integration/config] - 10https://gerrit.wikimedia.org/r/408522 (https://phabricator.wikimedia.org/T186587) (owner: 10Hashar) [15:15:12] (03Merged) 10jenkins-bot: debian-glue: run lintian as a hook [integration/config] - 10https://gerrit.wikimedia.org/r/408268 (https://phabricator.wikimedia.org/T186494) (owner: 10Hashar) [15:15:48] 10Continuous-Integration-Infrastructure, 10Release-Engineering-Team (Kanban), 10Packaging, 10Patch-For-Review: jenkins-debian-glue should run the lintian version from cowbuilder instead of from host - https://phabricator.wikimedia.org/T186494#3949108 (10hashar) a:03hashar [15:16:23] (03Merged) 10jenkins-bot: debian-glue: use lintian wikimedia profile [integration/config] - 10https://gerrit.wikimedia.org/r/408522 (https://phabricator.wikimedia.org/T186587) (owner: 10Hashar) [15:16:37] 10Continuous-Integration-Config, 10Packaging, 10Patch-For-Review: debian-glue lintian fails with: changes: bad-distribution-in-changes-file jessie-wikimedia - https://phabricator.wikimedia.org/T186587#3949114 (10hashar) 05Open>03Resolved a:03hashar That was an incomplete implementation of running linti... [15:17:30] 10Continuous-Integration-Infrastructure, 10Release-Engineering-Team (Kanban), 10Packaging, 10Patch-For-Review: jenkins-debian-glue should run the lintian version from cowbuilder instead of from host - https://phabricator.wikimedia.org/T186494#3945400 (10hashar) Left to do: rebuild a jenkins-debian-glue Deb... [15:19:18] 10Continuous-Integration-Infrastructure, 10Release-Engineering-Team (Kanban): npm-node-6-docker tests failing for Android project (test fallback to broken npm 1.4.21 instead of 3.8.3) - https://phabricator.wikimedia.org/T185931#3949135 (10hashar) //this comment is best seen at the web interface// antoine-approve [15:22:48] no_justification heh check out the projects view includes two links :) [15:23:20] and the changes include it too [15:23:21] https://gerrit.wikimedia.org/r/#/c/408540/ [15:27:44] PROBLEM - Puppet errors on deployment-redis01 is CRITICAL: CRITICAL: 100.00% of data above the critical threshold [0.0] [15:33:42] PROBLEM - Puppet errors on deployment-redis02 is CRITICAL: CRITICAL: 100.00% of data above the critical threshold [0.0] [15:35:08] 10Continuous-Integration-Infrastructure, 10Gerrit, 10Wikidata: Make DataTypes repository read-only - https://phabricator.wikimedia.org/T182566#3949190 (10hashar) And it got archived in CI via https://gerrit.wikimedia.org/r/#/c/402412/ [15:35:48] 10Gerrit, 10Patch-For-Review: Replace using certificates with tokens when using its-phabricator - https://phabricator.wikimedia.org/T178385#3949202 (10Paladox) 05Open>03Resolved [15:37:26] mutante no_justification i wonder what work needs to get done for this https://phabricator.wikimedia.org/T158915 to work? [15:38:36] 10Gerrit, 10Release-Engineering-Team (Someday), 10Operations: Make sure replying to emails in gerrit 2.14 works - https://phabricator.wikimedia.org/T158915#3949229 (10Paladox) p:05Lowest>03Normal Changing priority as we are now unlocked due to us updating to 2.14 :) [15:38:59] not the slightest idea :) i dislike email in general, the entire protocol :) [15:39:29] i want my browser-based notifications and i'm good [15:41:58] heh [15:43:13] PROBLEM - App Server Main HTTP Response on deployment-mediawiki05 is CRITICAL: CRITICAL - Socket timeout after 10 seconds [15:48:11] RECOVERY - App Server Main HTTP Response on deployment-mediawiki05 is OK: HTTP OK: HTTP/1.1 200 OK - 46805 bytes in 3.653 second response time [15:50:24] 10Continuous-Integration-Infrastructure, 10Zuul: Exception while launching job: TypeError: 'int' object has no attribute '__getitem__' - https://phabricator.wikimedia.org/T186381#3949302 (10hashar) That is what https://review.openstack.org/#/c/433748/ is supposed to fix. Namely Gerrit sends the change number a... [15:51:12] 10Continuous-Integration-Infrastructure, 10Zuul: Exception while launching job: TypeError: 'int' object has no attribute '__getitem__' - https://phabricator.wikimedia.org/T186381#3949305 (10Paladox) @hashar but now it's pointing at File "/usr/share/python/zuul/local/lib/python2.7/site-packages/zuul/model.p... [15:55:11] 10Continuous-Integration-Infrastructure, 10Zuul: Exception while launching job: TypeError: 'int' object has no attribute '__getitem__' - https://phabricator.wikimedia.org/T186381#3949333 (10Paladox) @hashar it seems change number is converted to a string in zuul, but patch number isen't. we had this problem in... [15:55:39] 10Continuous-Integration-Infrastructure, 10Zuul: Exception while launching job: TypeError: 'int' object has no attribute '__getitem__' - https://phabricator.wikimedia.org/T186381#3949335 (10Paladox) event.patch_number = patchset.get('number') > event.patch_number = str(patchset.get('number')) [16:10:40] 10Continuous-Integration-Infrastructure, 10Zuul: Exception while launching job: TypeError: 'int' object has no attribute '__getitem__' - https://phabricator.wikimedia.org/T186381#3949365 (10hashar) When a Depends-On header is set, Zuul queries Gerrit for open changes for the project which eventually invokes:... [16:12:28] 10Continuous-Integration-Infrastructure, 10Zuul: Exception while launching job: TypeError: 'int' object has no attribute '__getitem__' - https://phabricator.wikimedia.org/T186381#3949371 (10Paladox) record['currentPatchSet']['number'] should be str(record['currentPatchSet']['number']) too? and str(record['numb... [16:14:44] 10Phabricator: My username shows wrong - https://phabricator.wikimedia.org/T185998#3949376 (10CodeCat) To have my name show as Rua. [16:45:21] 10Phabricator: My username shows wrong - https://phabricator.wikimedia.org/T185998#3949495 (10Aklapper) p:05Triage>03Low a:05CodeCat>03Aklapper [16:45:48] 10Phabricator: Rename Phabricator user CodeCat to Rua - https://phabricator.wikimedia.org/T185998#3930496 (10Aklapper) [16:46:09] 10Phabricator: Rename Phabricator user CodeCat to Rua - https://phabricator.wikimedia.org/T185998#3930496 (10Aklapper) 05Open>03Resolved Done [16:48:35] uh, wikibase failing? https://integration.wikimedia.org/ci/job/mediawiki-extensions-hhvm-jessie/34810/console [16:53:18] https://en.wikipedia.beta.wmflabs.org/w/index.php?title=Selenium_Echo_link_test_0.7791232640420875&action=history < Nikerabbit related? [16:53:20] twentyafterfour: phab just broke [16:53:27] mutante: yup was about to say same [16:53:36] and i have a 1hr meeting which needs it haha :) [16:53:40] discussed in #wikimedia-operations too [16:55:16] mutante hes on vacation i think [16:57:24] and back. [16:58:40] restarting httpd fixed it [16:58:48] even though it looked like code issue first [16:58:50] jdlrobson: doesn't look related to me, but just guessing [17:01:10] Nikerabbit: yup pinned it down > https://phabricator.wikimedia.org/T186618 [17:01:34] Nikerabbit: actually i saw a similar problem to that error on MobileFrontend now i think about it [17:01:39] Nikerabbit: > https://gerrit.wikimedia.org/r/#/c/408547/ [17:01:44] not wikibase specific [17:01:51] but shows ResourcesTest::testFileExistence [17:02:12] rebasing made it go away [17:02:33] actually it's identical to your message :) [17:02:53] 10Release-Engineering-Team (Kanban), 10Release, 10Train Deployments: 1.31.0-wmf.20 deployment blockers - https://phabricator.wikimedia.org/T183959#3949549 (10Addshore) [17:03:35] mutante: mukunda is out this week [17:12:18] 10Release-Engineering-Team (Kanban), 10Release, 10Train Deployments: 1.31.0-wmf.20 deployment blockers - https://phabricator.wikimedia.org/T183959#3949570 (10matmarex) [17:12:35] jdlrobson: not able to rebase, they are already up to date [17:12:45] can you recheck ? [17:13:25] greg-g: gotcha, thanks. https://wikitech.wikimedia.org/wiki/Incident_documentation/20180206-Phabricator#Timeline .. WIP [17:14:00] luca will make a ticket [17:16:08] 10Continuous-Integration-Infrastructure, 10Operations, 10Patch-For-Review: legoktm can't deploy docker images on contint1001 - https://phabricator.wikimedia.org/T186475#3949603 (10Legoktm) @akosiaris thank you! Should we add the entire contint-admins team to contint-docker? I think people like @addshore are... [17:17:41] 10Phabricator, 10Release-Engineering-Team, 10Operations: Phabricator down due to - https://phabricator.wikimedia.org/T186620#3949613 (10elukey) p:05Triage>03High [17:17:59] 10Phabricator, 10Release-Engineering-Team, 10Operations: Phabricator down due to "Failed to `proc_open()`: proc_open() expects parameter 2 to be array" - https://phabricator.wikimedia.org/T186620#3949625 (10elukey) [17:18:03] 10Release-Engineering-Team (Kanban), 10Release, 10Train Deployments: 1.31.0-wmf.20 deployment blockers - https://phabricator.wikimedia.org/T183959#3949628 (10Addshore) [17:18:07] hello people [17:18:09] jdlrobson: yes, didn't help [17:18:16] anybody available for https://phabricator.wikimedia.org/T186620 ? [17:20:28] 10Continuous-Integration-Infrastructure, 10Operations, 10Patch-For-Review: legoktm can't deploy docker images on contint1001 - https://phabricator.wikimedia.org/T186475#3949638 (10Addshore) >>! In T186475#3949603, @Legoktm wrote: > @akosiaris thank you! Should we add the entire contint-admins team to contint... [17:22:31] 10Phabricator, 10Release-Engineering-Team, 10Operations: Phabricator down due to "Failed to `proc_open()`: proc_open() expects parameter 2 to be array" - https://phabricator.wikimedia.org/T186620#3949649 (10elukey) [17:22:43] 10Phabricator, 10Release-Engineering-Team, 10Operations: Phabricator down due to "Failed to `proc_open()`: proc_open() expects parameter 2 to be array" - https://phabricator.wikimedia.org/T186620#3949613 (10elukey) [17:24:18] 10Phabricator, 10Release-Engineering-Team, 10Operations: Phabricator down due to "Failed to `proc_open()`: proc_open() expects parameter 2 to be array" - https://phabricator.wikimedia.org/T186620#3949613 (10Dzahn) https://wikitech.wikimedia.org/wiki/Incident_documentation/20180206-Phabricator [17:26:23] no_justification wondering do you know how i can fix this mw error [17:26:23] mediawiki/oauthclient: 0.1.2 installed, ^0.1.2 required. [17:26:28] when running update.php [17:26:36] using the OAuth extension [17:26:41] please [17:27:31] Are you installing from master? Do composer update [17:27:40] no_justification nope from 1.30 [17:28:10] hmm, the OAuth extension dosen't even have mediawiki/oauthclient in composer.json [17:29:19] Maybe related T185618 [17:29:19] T185618: mediawiki/vendor repo has lost all its REL branches - https://phabricator.wikimedia.org/T185618 [17:29:39] ah [17:33:22] 10Continuous-Integration-Infrastructure, 10Operations, 10Patch-For-Review: legoktm can't deploy docker images on contint1001 - https://phabricator.wikimedia.org/T186475#3949700 (10akosiaris) >>! In T186475#3949638, @Addshore wrote: >>>! In T186475#3949603, @Legoktm wrote: >> @akosiaris thank you! Should we a... [17:34:21] sorry people, anybody available to talk about phabricator? It was returning 503s only until we restarted it [17:35:49] 10Continuous-Integration-Infrastructure, 10Operations, 10Patch-For-Review: legoktm can't deploy docker images on contint1001 - https://phabricator.wikimedia.org/T186475#3949702 (10Legoktm) >>! In T186475#3949700, @akosiaris wrote: >> I think people like @addshore are also going to need to be able to deploy d... [17:36:37] I'm gonna fix the mw/vendor thing today [17:38:27] * apergos eyes twentyafterfour [17:39:15] legoktm im wondering how do i install mediawiki/oauthclient from dev-master? [17:39:17] with composer [17:39:29] 10Continuous-Integration-Infrastructure, 10Operations, 10Patch-For-Review: legoktm can't deploy docker images on contint1001 - https://phabricator.wikimedia.org/T186475#3949716 (10akosiaris) >>! In T186475#3949702, @Legoktm wrote: >>>! In T186475#3949700, @akosiaris wrote: >>> I think people like @addshore a... [17:39:47] i think it's complaning that i carn't use version 0.1.2 but i can use a newer release. [17:42:59] 10Phabricator, 10Release-Engineering-Team, 10Operations, 10User-Elukey: Phabricator down due to "Failed to `proc_open()`: proc_open() expects parameter 2 to be array" - https://phabricator.wikimedia.org/T186620#3949723 (10elukey) [17:43:20] found it [17:43:25] composer require mediawiki/oauthclient:dev-master [17:48:41] PROBLEM - Free space - all mounts on deployment-tin is CRITICAL: CRITICAL: deployment-prep.deployment-tin.diskspace._mnt.byte_percentfree (No valid datapoints found)deployment-prep.deployment-tin.diskspace._srv.byte_percentfree (<10.00%) [18:06:19] 7.6G mediawiki-staging [18:06:50] 1.8G analytics [18:06:50] 4.0G ores [18:07:31] 304M parsoid [18:07:31] 314M mathoid [18:07:31] 394M electron-render [18:07:31] 399M ocg [18:07:31] 402M 3d2png [18:07:50] not sure what can really be cleaned up, there's a lot more directories with presumably just copies of their repos [18:12:56] one thing we do in beta that is still behind a feature flag in production is the /srv/mediawiki/.git repo that gets large over the period of a month or so (which is why it's still behind a feature flag :)) [18:13:27] removing that directory clears space, and it will be regenerated on the next deployment so it's harmless. [18:14:25] !log removing /srv/mediawiki/.git on deployment-tin to clear space [18:14:31] Logged the message at https://wikitech.wikimedia.org/wiki/Release_Engineering/SAL [18:23:35] ah thanks [18:24:18] I wonder if it's worth a cron job... meh [18:27:23] no_justification heh the change list in poly was just updated http://gerrit-new.wmflabs.org/r/q/status:open :) [18:28:41] RECOVERY - Free space - all mounts on deployment-tin is OK: OK: deployment-prep.deployment-tin.diskspace._mnt.byte_percentfree (No valid datapoints found) [18:30:54] thcipriani: you around? [18:31:15] arlolra: yep [18:31:35] Do you have time for a question? [18:31:41] sure, what's up? [18:32:27] In the middle of my deploy, my ssh connection was cut and I wasn't using a screen session [18:33:08] How do I recover from, [18:33:10] "Failed to acquire lock "/tmp/scap.parsoid.lock"; owner is "arlolra"; reason is ..." [18:35:02] arlolra: if you remove that file manually it should let you run a new deployment [18:35:45] rm /tmp/scap.parsoid.lock that is [18:36:11] right [18:36:17] seems to be doing the trick [18:36:28] cool :) [18:36:30] thanks [18:36:36] glad I could help! [18:39:24] PROBLEM - Puppet errors on deployment-mx is CRITICAL: CRITICAL: 100.00% of data above the critical threshold [0.0] [18:50:56] 10Phabricator (Upstream), 10GitHub-Mirrors, 10Epic, 10Upstream: Create system to sync GitHub issues with Phabricator - https://phabricator.wikimedia.org/T86991#981141 (10Tgr) Simpler alternative: * Write [[https://github.com/blog/1184-contributing-guidelines|CONTRIBUTING.md]] explaining that bugs need to b... [18:55:54] 10Continuous-Integration-Infrastructure, 10Operations, 10Patch-For-Review: legoktm can't deploy docker images on contint1001 - https://phabricator.wikimedia.org/T186475#3950009 (10hashar) Both @Legoktm and @Addshore already have the privileges to run privileged code on CI and can really run any random Docker... [18:57:48] legoktm or no_justification hi, could you review https://gerrit.wikimedia.org/r/408578 please :) [19:00:04] 10Gerrit, 10Operations, 10Patch-For-Review, 10Performance: New gerrit login ui is causing performance problems when going through gerrit.wikimedia.org - https://phabricator.wikimedia.org/T185506#3950028 (10Krinkle) @demon I understand. My point is merely about the "flash of unstyled content", which is actu... [19:17:14] CI seems to have got stuck again, this time on 408295. [19:17:57] James_F i guess this is the same as https://phabricator.wikimedia.org/T186381 [19:18:09] cc hashar legoktm ^^ [19:19:01] most probably yup [19:21:58] I removed the C+2 but it didn't change anything. [19:22:28] yeah that doesn't dequeue the patch from the pipeline [19:22:45] I am thinking of a way to do it [19:23:01] maybe edit the commit summary to create a new patchste and +2 it [19:23:07] it might then discard the older patchset [19:23:13] trying [19:24:21] hmm no [19:25:18] !log Restarted Zuul due to T186381 [19:25:24] Logged the message at https://wikitech.wikimedia.org/wiki/Release_Engineering/SAL [19:25:24] T186381: Exception while launching job: TypeError: 'int' object has no attribute '__getitem__' - https://phabricator.wikimedia.org/T186381 [19:26:43] 10Continuous-Integration-Infrastructure, 10Zuul: Exception while launching job: TypeError: 'int' object has no attribute '__getitem__' - https://phabricator.wikimedia.org/T186381#3942743 (10Jdforrester-WMF) Happened again, this time with 408295. Both by me, but otherwise unconnected (different repos), so not s... [19:26:49] will come back in half an hour or so [19:27:06] thcipriani: so in short it seems any change that enters gate and submit and have a depends-on end up being stuck [19:27:12] I am tempted to go with a dirty fix [19:27:24] reading some stories to kids and I will be back in half an hour [19:27:25] that seems bad. [19:27:30] hashar: Ah. That's the common thread. [19:27:31] James_F: yeah changes with a depends-on are stuck [19:27:37] :-( [19:27:53] hashar it could be because patch num is a int [19:27:55] Zuul has a fault when it look up the depends-on change from Gerrit [19:28:00] as is not being converted to a string [19:28:05] (where as change num is) [19:28:20] gerrit yield an integer for the change number and then zuul tries to run string methods on it => error! [19:28:31] Yep [19:28:40] tyler and I hve been talking about it a couple hours ago and we get an idea for a good fix [19:28:52] anyway I will be back soon [19:28:56] Cool. [19:31:49] James_F: are you sure https://gerrit.wikimedia.org/r/#/c/408526/ is right? [19:32:03] (first time I do something like that) [19:32:50] Hauskatze: It looks right to me. A lot of the RevisionStore stuff is new, though. [19:33:19] okay :-) [19:34:42] zuul is not picking the change for some reason [19:36:24] Zuul just got restarted. No panic. [19:38:38] ok [19:40:09] I'll have to rebase https://gerrit.wikimedia.org/r/#/c/406280/ after 408526 gets merged [19:40:38] 10MediaWiki-Codesniffer: New Sniff for using pointless variables before return - https://phabricator.wikimedia.org/T179768#3735439 (10Umherirrender) Just for the record: This has to check, if the variable is never used before, because some cache results are set directly before a return[1] It should not the funct... [19:46:16] 10MediaWiki-Codesniffer: License "GPL-2.0+" is a deprecated SPDX license identifier, use "GPL-2.0-or-later" instead - https://phabricator.wikimedia.org/T186481#3945182 (10Umherirrender) I would assume this is true for all composer packages from wikimedia This makes it harder for a security release, because it h... [19:58:02] 10Continuous-Integration-Infrastructure, 10Zuul: Exception while launching job: TypeError: 'int' object has no attribute '__getitem__' - https://phabricator.wikimedia.org/T186381#3950224 (10hashar) [19:59:21] 10Continuous-Integration-Infrastructure, 10Zuul: Exception while launching job: TypeError: 'int' object has no attribute '__getitem__' - https://phabricator.wikimedia.org/T186381#3942743 (10hashar) That happens whenever a change having Depends-On enters gate-and-submit. Zuul does a query to Gerrit to find whet... [20:08:17] (03PS1) 10Tpt: Adds a dependency on Wikibase to Wikisource extension [integration/config] - 10https://gerrit.wikimedia.org/r/408599 [20:14:06] I am seeing a bunch of these: 19:51:48 PHP Notice: Undefined index: maintenance/edit.php in /srv/jenkins-workspace/workspace/mediawiki-phpunit-coverage-patch/src/vendor/legoktm/clover-diff/src/Diff.php on line 87 [20:14:06] in the CI logs. I assume this is from new coverage patch? Is it normal? [20:14:27] legoktm: ^^^ [20:15:04] 10Gerrit: Support OAuth for login onto gerrit.wikimedia.org - https://phabricator.wikimedia.org/T147864#3950275 (10Tgr) It's not a very good standard though (e.g. the registration experience we provide is pretty bad) and probably probably contributes to our dwindling volunteer numbers. Plus having passwords is a... [20:17:33] 10Release-Engineering-Team, 10phpunit-patch-coverage: phpunit-patch-coverage CI task produces lots of notices - https://phabricator.wikimedia.org/T186638#3950282 (10Smalyshev) [20:30:43] SMalyshev: yeah, it's a bug. but you can ignore them, it doesn't affect the output [20:31:07] legoktm: ok, I filed a phab task, I'll ignore them otherwise [20:35:35] legoktm: https://gerrit.wikimedia.org/r/#/c/406332/ ? :) [20:35:56] (03PS1) 10Umherirrender: Avoid assignment in return statements [tools/codesniffer] - 10https://gerrit.wikimedia.org/r/408610 (https://phabricator.wikimedia.org/T170332) [20:37:17] Hauskatze: commented [20:37:29] * Hauskatze checks [20:38:23] legoktm: and what about using manifest_version 2? [20:38:45] unless you want to increase compat [20:38:59] PROBLEM - Puppet errors on deployment-videoscaler01 is CRITICAL: CRITICAL: 100.00% of data above the critical threshold [0.0] [20:39:01] PROBLEM - Puppet errors on deployment-mediawiki07 is CRITICAL: CRITICAL: 100.00% of data above the critical threshold [0.0] [20:39:10] 10Continuous-Integration-Infrastructure, 10Zuul: Exception while launching job: TypeError: 'int' object has no attribute '__getitem__' - https://phabricator.wikimedia.org/T186381#3950330 (10hashar) I am hunting the code path used by Zuul when it processes the json. My plan to cast it to a string as close as po... [20:45:35] PROBLEM - Puppet errors on deployment-tmh01 is CRITICAL: CRITICAL: 100.00% of data above the critical threshold [0.0] [20:46:54] 10MediaWiki-Codesniffer: Sniff to $wmf* -> $wmg* - https://phabricator.wikimedia.org/T186642#3950368 (10MarcoAurelio) [20:55:22] 10Gerrit, 10Patch-For-Review: Setup gitiles for use with gerrit as a repo viewer - https://phabricator.wikimedia.org/T184116#3950416 (10Dzahn) cool, this ticket seems resolved [20:55:57] 10Gerrit, 10Patch-For-Review: Setup gitiles for use with gerrit as a repo viewer - https://phabricator.wikimedia.org/T184116#3950418 (10Dzahn) I think the answer is "no separate domain"? [21:11:32] it seems all patches on zuul are being linked one to another, so until the last patch is not checked, no one gets merged; is that so? [21:12:01] 408090,4 is not being tested [21:12:06] hashar legoktm ^^ [21:12:10] sigh [21:12:12] so it's blocking the rest [21:12:16] do I need ot restart it again [21:12:21] cf gate-and-submit [21:12:31] all are being marked as dependencies [21:12:35] :| [21:12:37] legoktm i am not sure if this is the same problem though [21:12:44] i see it not depending on any other changes [21:12:45] Hauskatze: there's a zuul bug in which Depends-On sometimes causes the whole thing to break [21:12:50] I see 2 depends-on [21:12:55] on 408090,4 ? [21:13:03] yeah... [21:13:03] https://gerrit.wikimedia.org/r/#/c/408090/ [21:13:04] oh [21:13:33] butler jenkins needs a couple of footmen to help :P [21:14:11] !log restarted zuul due to patch being stuck (T186381) [21:14:27] Logged the message at https://wikitech.wikimedia.org/wiki/Release_Engineering/SAL [21:14:27] T186381: Exception while launching job: TypeError: 'int' object has no attribute '__getitem__' - https://phabricator.wikimedia.org/T186381 [21:15:31] (03Draft1) 10Paladox: Fix support for gerrit 2.14+ [integration/zuul] (patch-queue/debian/jessie-wikimedia) - 10https://gerrit.wikimedia.org/r/408625 [21:15:33] (03PS2) 10Paladox: Fix support for gerrit 2.14+ [integration/zuul] (patch-queue/debian/jessie-wikimedia) - 10https://gerrit.wikimedia.org/r/408625 [21:15:44] legoktm i wonder if that ^^ will fix it [21:15:48] thcipriani: turns out it is a little bit more complicated than just blindly casting to string :( [21:16:05] the queue already got stuck agian on https://gerrit.wikimedia.org/r/#/c/408385/-1..2 [21:16:13] because they treat it as an int somewhere else? [21:16:24] yeh [21:16:25] 10Gerrit, 10Operations, 10Patch-For-Review, 10Performance: New gerrit login ui is causing performance problems when going through gerrit.wikimedia.org - https://phabricator.wikimedia.org/T185506#3950466 (10Krinkle) p:05Triage>03Low a:05Paladox>03Krinkle [21:16:35] apparently only if they use Depends-On: [21:16:54] thcipriani: I will go with your @property thingy :] [21:17:28] hashar would something like https://gerrit.wikimedia.org/r/#/c/408625/2/zuul/connection/gerrit.py that fix it? :) [21:17:32] yeah, @property with @number.setter seems sane to me if it's always meant to be a string. [21:17:42] paladox: no [21:17:47] oh [21:17:58] paladox: it is wrong on multiple level. That is the change number which is affected, not the patchset number [21:18:06] oh i see [21:18:57] (03Abandoned) 10Paladox: Fix support for gerrit 2.14+ [integration/zuul] (patch-queue/debian/jessie-wikimedia) - 10https://gerrit.wikimedia.org/r/408625 (owner: 10Paladox) [21:20:08] I think someone is oging to have to restart it again [21:20:13] but I'm going afk for class now [21:20:17] good luck :) [21:20:45] oh actually I am stupid [21:21:03] thcipriani: since self.number[-2:] has an issue because self.number is an int [21:21:08] we could cast a string right there :] [21:22:17] sure, but that's potentially whack-a-mole if there are other places self.number is treated as a string since they updated one place where they build the Changeish object to be change.number = str(number) and not the other place. [21:22:34] although that would fix our immediate issue. [21:23:17] I have no qualms with either fix. [21:28:39] tests.test_model.TestChangeish.test_getBasePath_cast_change_number_to_string 0.001 [21:28:40] PASSED [21:29:19] I started looking at hacking the source of the json [21:29:31] and tweak the json sent by Gerrit. Even had a few tests for that [21:29:39] but that is probably too terrible [21:33:46] how's Jenkins doing? [21:34:28] (03PS1) 10Hashar: wmf: change number must be a string when formatting [integration/zuul] (patch-queue/debian/jessie-wikimedia) - 10https://gerrit.wikimedia.org/r/408630 (https://phabricator.wikimedia.org/T186381) [21:35:22] (03CR) 10Thcipriani: [C: 031] wmf: change number must be a string when formatting [integration/zuul] (patch-queue/debian/jessie-wikimedia) - 10https://gerrit.wikimedia.org/r/408630 (https://phabricator.wikimedia.org/T186381) (owner: 10Hashar) [21:37:19] Volker_E zuul currently has a bug see https://phabricator.wikimedia.org/T186381 [21:37:35] 408385,2 [21:37:37] is stuck [21:37:37] thanks paladox [21:37:46] your welcome [21:38:26] jdlrobson hi, to merge https://gerrit.wikimedia.org/r/#/c/408385/ you need to remove the depends on [21:38:33] hashar: / thcipriani - can you re+2 https://gerrit.wikimedia.org/r/#/c/406280/ please? [21:38:37] due to a bug https://phabricator.wikimedia.org/T186381 [21:38:44] that is currently being fixed by hashar :) [21:39:03] thanks paladox for the heads up [21:39:09] your welcome :) [21:39:47] though i think zuul needs to be restarted now [21:40:36] hashar zuul needs restarting to remove 408385,2 please [21:40:51] (03PS1) 10Hashar: 2.5.1-wmf2: fix ChangeIsh.basePath [integration/zuul] (debian/jessie-wikimedia) - 10https://gerrit.wikimedia.org/r/408686 (https://phabricator.wikimedia.org/T186381) [21:41:05] !log Rebuilding Zuul package to hotfix T186381 [21:41:14] Logged the message at https://wikitech.wikimedia.org/wiki/Release_Engineering/SAL [21:41:15] T186381: Exception while launching job: TypeError: 'int' object has no attribute '__getitem__' - https://phabricator.wikimedia.org/T186381 [21:42:12] Volker_E: yeah it is stuck. Will be fixed soonish [21:42:36] hashar: I've subscribed to the task, guess, when it's resolved I can continue [21:42:44] hashar: thanks! [21:45:16] and I forgot to mail wikitech :( [21:45:59] 10Continuous-Integration-Infrastructure, 10Patch-For-Review, 10Zuul: Exception while launching job: TypeError: 'int' object has no attribute '__getitem__' - https://phabricator.wikimedia.org/T186381#3950560 (10hashar) https://people.wikimedia.org/~hashar/debs/zuul_2.5.1-wmf2/ $ debdiff zuul_2.5.1-wmf1_amd64... [21:49:24] Project mwext-phpunit-coverage-publish build #651: 04FAILURE in 22 sec: https://integration.wikimedia.org/ci/job/mwext-phpunit-coverage-publish/651/ [21:49:48] Yippee, build fixed! [21:49:49] Project mwext-phpunit-coverage-publish build #652: 09FIXED in 24 sec: https://integration.wikimedia.org/ci/job/mwext-phpunit-coverage-publish/652/ [21:49:54] upgrading zuul [21:52:06] James_F: would you mind CR+2 the oojs/ui changes https://gerrit.wikimedia.org/r/#/c/401422/ and https://gerrit.wikimedia.org/r/#/c/408545/ ? They got flushed when I restarted Zuul [21:52:54] hashar: Done. [21:53:03] (03CR) 10Paladox: [C: 031] 2.5.1-wmf2: fix ChangeIsh.basePath [integration/zuul] (debian/jessie-wikimedia) - 10https://gerrit.wikimedia.org/r/408686 (https://phabricator.wikimedia.org/T186381) (owner: 10Hashar) [21:53:17] hashar: Does this mean depends-on patches should merge now? [21:54:21] James_F: also https://gerrit.wikimedia.org/r/#/c/406280/ if possible [21:54:30] James_F: hopefully. if you have an example I am willing to try [21:54:33] else I will forge one [21:54:53] hashar: Yep, definitely have one. :-) [21:54:59] hashar seems it is working [21:55:01] which pair ? [21:55:04] hashar https://gerrit.wikimedia.org/r/#/c/408690/ [21:55:20] the issue happens whenever both changes enter the gate and are both open [21:55:53] hashar: 408090 in Cite and 408295 in VE-MW. Oh, but that one's now merged. [21:56:01] ;D [21:56:20] hashar: But the one in Cite has a Depends-On on it and needs merging. OK for me to sling it in? [21:56:27] let me verify [21:57:12] James_F: yeah the Cite patch https://gerrit.wikimedia.org/r/#/c/408090/ would have worked previously [21:57:16] since both depends on got merged [21:57:40] +2ed [21:59:08] Ta. [22:00:24] paladox thcipriani: I think that did it [22:00:29] hashar :) [22:00:30] will craft an incident report tomorrow [22:00:43] should we forward the fix upstream? [22:03:21] https://github.com/openstack-infra/zuul/blob/3f3b74426363b45afc1493aeab629a3a00d37f01/zuul/driver/gerrit/gerritmodel.py#L45 [22:05:57] hashar lol they reverted the change by looks of it in zuul v3 (they didnt merge the change into v3) [22:05:58] https://github.com/openstack-infra/zuul/blob/4e24f868b25bbcadb642772acb3924f8864398ff/zuul/model.py#L2026 [22:07:35] 10Continuous-Integration-Infrastructure, 10Release-Engineering-Team (Kanban), 10Zuul: Exception while launching job: TypeError: 'int' object has no attribute '__getitem__' - https://phabricator.wikimedia.org/T186381#3950611 (10hashar) p:05Unbreak!>03High a:03hashar Posted a quick note on wikitech-l. I... [22:09:17] paladox: that depends on how the change.number property is set [22:09:23] oh [22:09:30] there is a nicer patch that can be made. Tyler suggested using an @property on the ChangeIsh class [22:09:36] so that when one does either of: [22:09:39] change.number = 5 [22:09:42] oh i see [22:09:46] or change.number = '42' [22:10:03] we can have it normalized to either integer or string [22:10:20] there are a bunch of tests that have to be adjusted as well [22:10:31] yep [22:10:58] hashar apparently they changed it in v3 [22:11:02] they get it from the url [22:11:05] per clarkb [22:11:07] :) [22:19:28] \o/ [22:20:05] (03CR) 10Hashar: [V: 032 C: 032] wmf: change number must be a string when formatting [integration/zuul] (patch-queue/debian/jessie-wikimedia) - 10https://gerrit.wikimedia.org/r/408630 (https://phabricator.wikimedia.org/T186381) (owner: 10Hashar) [22:20:31] (03CR) 10Hashar: [C: 032] 2.5.1-wmf2: fix ChangeIsh.basePath [integration/zuul] (debian/jessie-wikimedia) - 10https://gerrit.wikimedia.org/r/408686 (https://phabricator.wikimedia.org/T186381) (owner: 10Hashar) [22:22:09] (03Merged) 10jenkins-bot: 2.5.1-wmf2: fix ChangeIsh.basePath [integration/zuul] (debian/jessie-wikimedia) - 10https://gerrit.wikimedia.org/r/408686 (https://phabricator.wikimedia.org/T186381) (owner: 10Hashar) [22:23:21] guess it is fixed [22:23:25] I am going to bed [22:40:46] Woo! I like the new "browse" links in Gerrit [22:40:58] They integrate directly with Gitiles [22:42:13] Krinkle yep [22:42:14] Now all we need is for either the difussion or Gitiles viewer to support viewing PNGs and SVGs [22:42:25] Krinkle gitiles will never support those [22:42:27] though [22:42:37] Upstream blocked it based on security concerns [22:43:08] Yeah, but that's a bit lame. Doing it the wrong way is insecure yes, but there is a right way, and they're not priotising the implementation of that. [22:43:29] GitLab and GitHub did it years ago, in a secure way. [22:43:53] Krinkle https://gerrit-review.googlesource.com/#/c/gitiles/+/107850/ https://gerrit-review.googlesource.com/#/c/gitiles/+/135471/ https://gerrit-review.googlesource.com/#/c/gitiles/+/78140/ [22:43:57] Phabricator has the support already built-in, but not yet exposed for Difussion. [22:44:07] Phab does it by serving from an alternate domain [22:44:29] Krinkle polygerrit allows you to view images anyways (now) :) [22:44:52] GitHUb does it the same way as Phab, alternate domain. Which makes it secure essentially. [22:45:10] paladox: Thx for the links [22:45:17] Your welcome :) [22:47:25] damn, i just wrote a comment under my real name (*SIGH*) [22:55:24] Krinkle https://github.com/google/gitiles/issues [22:55:30] apparently issues get filled @ ^^ [22:57:32] https://github.com/google/gitiles/issues/7 [23:01:40] Krinkle would this work for us https://gerrit-review.googlesource.com/#/c/gitiles/+/78140/ [23:02:18] (note that the owner of that change is no longer alive, died a few weeks ago, so someone else will have to take over the change) [23:04:05] paladox: Sounds like the same approach that Phabricator takes, possibly even better (by allowing recovery for an invalid token, e.g. when sharing links) [23:04:06] Very nice [23:04:13] :) [23:04:16] I will rebase it then [23:04:31] Krinkle (also that was gerrit's founder who created that change) :) [23:10:08] Krinkle rebased it :) [23:17:56] Project mwext-phpunit-coverage-publish build #657: 04FAILURE in 39 sec: https://integration.wikimedia.org/ci/job/mwext-phpunit-coverage-publish/657/ [23:18:12] 10Gerrit, 10Release-Engineering-Team (Someday), 10Operations: Make sure replying to emails in gerrit 2.14 works - https://phabricator.wikimedia.org/T158915#3950905 (10demon) p:05Normal>03Low Actually I think this is kinda low priority since we don't offer the service yet--it would be an enhancement. [23:24:17] 10Project-Admins: WikiIndaba conference 2018 - https://phabricator.wikimedia.org/T186660#3950926 (10Dyolf77) [23:24:48] 10Gerrit, 10Patch-For-Review: Setup gitiles for use with gerrit as a repo viewer - https://phabricator.wikimedia.org/T184116#3950939 (10demon) Problem with second domain is that I dunno how well it'd work with being logged in (gitiles respects ACLs, one of the reasons I wanted to go the plugin route instead of... [23:28:30] 10Gerrit: Support OAuth for login onto gerrit.wikimedia.org - https://phabricator.wikimedia.org/T147864#3950944 (10demon) 05declined>03stalled That's totally a route to go too....OAuthing services that currently are behind LDAP would be awesome. But I figure it's better to be consistent with other LDAP-backe... [23:32:14] Krinkle apparently we will have to wait for 2.16 / 3.0 for it. [23:34:44] Looks like Zuul 'test' pipeline is stalled with lots of changes that have already been merged meanwhile [23:34:49] Would be nice if we could dequeue those somehow [23:35:03] Looks like zuul doesn't even have a cli option for it to do manually [23:46:40] 10Gerrit: Support OAuth for login onto gerrit.wikimedia.org - https://phabricator.wikimedia.org/T147864#3950975 (10Tgr) Could probably use mod_authnz_fcgi with something like mediawiki/oauthclient? Anyway this is blocked on having some connection between LDAP and SUL (so {T148048} I guess?). [23:55:03] (03CR) 10Jforrester: [C: 031] Avoid assignment in return statements [tools/codesniffer] - 10https://gerrit.wikimedia.org/r/408610 (https://phabricator.wikimedia.org/T170332) (owner: 10Umherirrender) [23:56:53] 10Release-Engineering-Team (Kanban), 10MediaWiki-Cache, 10MediaWiki-Vagrant, 10MW-1.31-release-notes (WMF-deploy-2018-02-06 (1.31.0-wmf.20)), and 3 others: "User should be able to change preferences" Selenium test fails when targeting mediawiki-vagrant - https://phabricator.wikimedia.org/T185328#3950987 (10...